Did a small blockchain startup company just solve a major problem in the academic peer review process?

Academic science runs by rules which are often opaque to outsiders. To understand one of the biggest problems in the academic peer review process, here a short introduction first. If you know about this already, you might want to jump to the next part.

How publishing and peer review works (in a nutshell)

When you are a scientist, you need to publish. In academic research, “papers” are the visible currency of our work. And like any other currency, some work has more value than other.

How much your chunk of work (summarised in a paper) is worth is mainly judged by a simple factor: in which journal is the publication published? This system is far from perfect. In fact, its utterly broken since the prestige (value) that is attached to a journal can be measured by several metrics; the most widely used is the impact factor.

To be able to have your manuscript accepted in a particular journal, a manuscript is submitted by the scientists to an editor of that journal. The editor decides then if the manuscript is of potential interest and if it fits to the scope of the journal at hand. If the answer is yes (in most cases, the answer is no, btw), the editor sends the manuscript to 2-5 other scientist in the field. These “peers” of the authors will individually review the paper anonymously and send their comments back to the editor together with a recommendation if the paper should be accepted directly (virtually never happens), if the text needs to be improved (aka “minor revision”; happens always), if the scientific work is not complete yet and additional experiments/analyses need to be added to the the manuscript (“major revision”; happens often) or if the quality of the work is not sufficient and should be rejected. The latter will constitute for the majority of cases. If a manuscript is rejected, the authors will usually submit it to another journal. And the cycle starts from the beginning.

So, whats the problem here? First of all, peer review is necessary to ensure quality of scientific work. The fact that the reviewers can review the manuscript anonymously also means that even the most powerful scientists can be criticised without fear of retaliation. Still, the complete system is fundamentally broken; diverse studies (and empirical data accumulated by every scientist throughout his/her professional life) show that peer review is not performing well. A complete website is dedicated to track papers that need to be retracted; often after decades of being published. Don’t go there to check it out. Its depressing.

The (biggest?) problem for scientists in the peer review system

Scientists are part of this system. To be able to critically review a manuscript from a colleague, you need years of training and experience. Only active scientists working in the same field of research will be able to provide this level of expertise usually. Funnily, we are doing this work for free; scientists are not paid or in any other form compensated for their work for publishers*. Reading a manuscript and write a review is surprisingly tedious work. Dumping a complete days work into a review is rather the norm.
This means that each manuscript which is send out will keep 3 other scientists busy. I do not know a single researcher who is not complaining about the huge workload coming from writing peer reviews. To make things worse, every rejected manuscript will add another 2 or 3 reviews since the first reviews are only known to the editor of one journal. To make this clear: every rejection means that the work of 3 highly qualified experts is dumped and left unused. Each rejection adds more and more reviews back to the shoulders of the scientific community. Did I say already we are doing this for free?

Blockchain to publishing

How could the situation be improved? Lowering standards to avoid rejections of low quality manuscripts cannot be the answer. In fact, so much poorly conducted research is published already that science is in danger to loose it legitimacy before the public.
One solution could come from an unexpected side: the blockchain. I am not qualified to explain what a blockchain really is. Others are, however. For the sake of this article, lets say that a blockchain is a decentralised database that has a special feature: its content is inmutable. The blockchain is distributed over thousands of computers (nodes) which continuously validate new entries into the database in between each other. Once started, each node is up-to-date with all information in the database: a fully decentralised system.
Blockchain-based technologies are receiving a lot of hype in these days and many tech startups are challenging established companies by “blockchainifying” services. Estonia (yes, the baltic country in the EU) thinks about migrating its complete digital infrastructure to the blockchain.

Publica.io

Blockchain startup companies have developed a new way to find venture capital. In 2017, hundreds of initial coin offerings (ICOs) were kicked off. These startups sell digital coins (or tokens) which will grant access to their blockchain or service. One of these startups is Publica.io, a small Lithuanian startup. Their mission is to use the blockchain for decentralised publishing. And while the company is for now focusing on independent authors and starting their business, they might just have solved the problem for scientific peer review.

Decentralised peer review

If we quickly recapitulate the problem I addressed above (too many reviews are lost because they never leave the desk of individual editors and publishers), its immediately obvious, how a blockchain could help: if a manuscript is no longer submitted to single journals but added to an independent decentralised peer review blockchain, the very same reviews that are lost today could be easily made “sticky” to the manuscript. If a rejected manuscript is now send to another journal, the editors and reviewers could directly follow the changes that were implemented in the manuscript to address those first reviews. A history of the complete peer review process would suddenly be possible. I am very sure that this would help to significantly reduce the workload on scientists from reviews and could speed up decisions if a manuscript is accepted tremendously since the complete review process does not always start from scratch again.

And else? A double-blind review process should not be a problem with such a system (if that is useful is another discussion). Also, a decentralised “paper blockchain” could gather all paper metrics which are nowadays often only found on the journals websites.

Finally: changing something in scientific publishing is incredibly hard. The publishers have a well-oiled money machine running and no incentives to give anything out of their hands. However, they need our expertise and our work. The blockchain could impact the life of every scientist world-wide: by reducing their work and speeding up decisions on papers. Sounds interesting?

---

* I will never understand why tax payers are allowing this, but thats another story.

---

Full disclosure statement: I am not affiliated with publica.io or have been compensated in any form to write for them. I participated however in the publica.io ICO in November 2017.